You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.


Oil spill detection on the Galician coast using a neuro-symbolic framework and ASAR images

Jesus M. Torres Palenzuela(1) , Luis Gonzalez Vilas(1) , and Juan M. Corchado(2)

(1) University of Vigo, Facultad de Ciencias, Campus Lagoas-Marcosende, 36200,Vigo, Spain
(2) University of Salamanca, Facultad de Ciencias, 37008, Salamanca, Spain


The Prestige tanker caused an extensive oil spill off the north-west coast of Spain on November 2002, occasioning a big environmental and economic catastrophe in the Galician region. The Spanish project CONTINMAR is aimed to the design of a contingency plan in the event of accidental marine pollution on the area, including the development of an oil spill detection, monitoring and forecasting system using remote sensing and artificial intelligence techniques. Radar imagery have proven a very useful tool in order to detect oil spills due to the dampening effect of oil on water, and different classification algorithms have been proposed. A hybrid artificial intelligence approach to the problem of classification of oil spills in marine environments offers potential advantages over alternative systems, because it is able to deal with uncertain, incomplete and even inconsistent data. The final classification system that is proposed in this paper will be a CBR-based neuro-symbolic framework, which will integrate and include all the hybrid artificial intelligence techniques and methods investigated in the framework of this project to determine the best possible the characteristics of the detected oil spills. The feasibility of the classification system will be analysed starting from historical data arisen from ENVISAT ASAR images acquired during the Prestige catastrophe, between November 2002 and April 2003.



  Higher level                 Last modified: 07.10.03